JUNE
1058
NOTES
Acknowledgment. The authors wish to thank Emery Industries, Inc., for a generous fellowship (J.D.F.) :md Moshe Ish-Shalom for valuable technical assistance. APPLIEDSCIEUCE DEPARTMEKT UXIVERSITY OF CINCIXNATI CINCI~\ATI, OHIO
Isolation of an Alkaloid, Annuloline, from the Roots of Lolium M u l t i J l ~ r r c m ’ ~ ~ BERUARD AXELRODS A N D J. REXEBELZILE‘ Received December 9, 1957
The roots of the annual rye grass (Lolium multi-
$orurn) have been known for some time to contain a brilliant blue fluorescing pigments6Seed analysts are able to distinguish annual rye grass seed from the morphologically similar perennial variety (L. perennc) by viewing the roots of the 8-day old seedlings under long wave ultraviolet light; the perenne roots do not fluoresce.6 The extraction of the fresh roots of the seedlings obtained from about 100 Ibs. of seed of L. multi$orurn has enabled us to isolate and crystallize the fluorescent principle which appears to be a weakly basic alkaloid. The tentative empirical formula, C2,,HI9T\TO4,best satisfies the presently available data. It seems likely that there are three methoxyl groups present per formula weight. A variety of organic solvents such as ethanol, methylene chloride, carbon tetrachloride, ether, benzene, arid petroleum ether serve to extract this substance. Since it is both an extremely weak base and only sparingly soluble in water it cannot be readily extracted with hydrochloric acid. Having no acidic (or phenolic) groups it is not extractable with alkali. The pigment is strongly fluorescent. I n petroleum ether it fluoresces with an intensity about 13 times that of quinine sulfate in 0.1 N sulfuric acid on a weight basis when compared in a model 12B Colerrtan fluorophot,ometer fitted with B-1 and PC-1 filters. It can be readily detected on filter paper by virtue of its fluorescence. I n general it exhibits high mobilities on being chromato(1) Journal Paper No. 1205 of the Purdue Agricultural Experimeiit Station, Purdue University, W. Lafayette, Ind. ( 2 ) A portion of this work is taken from the thesis submitted by J. R. B. in partial fulfillment of the requirements for the ill S. Degree, Purdue University, August, 1957. (3) To whom inquiries should be sent. (4) Present Address: Dept. of Chemistry, Ontario Agricultural College, Guelph. (5) G. Gentner, Prakt. Blatter Pjlanzenbau u. PfEanzenschutz, 6, 166 (1929). (6) U. S. Ilept. of Agriculture, Testing Agricztltural and Vegetable Seeds, Government Printing Office, Washington, D. C. (l952), pp. 103-104.
918
graphed on paper with solvents of lorn water content. Acidification of solvents with hydrochloric acid or basification with ammonium hydroxide such as employed by Swain7 as diagnostic procedures for discriminating between acidic (or enolic) and basic plant pigments did not influence the mobility of the pigment. A hydrochloride can be obtained by the action of anhydrous hydrogen chloride on a solution of the alkaloid in ethanol or petroleum ether. The resulting salt is strongly acidic. Because the characteristic blue fluorescence of solutions of the alkaloid is not shown by the salt form it is possible to estimate from a plot of fluorescence vs. pH that the base is approximately 50% in the free form a t pH 2.0. By means of ultraviolet light as little as 0.01 mcg. of the alkaloid may be detected when applied to a 2-mm. diameter area on Whatman No. 1 filter paper. Exposure of the spot to fumes of hydrochloric acid abolishes the blue fluorescence and the spot now glows yellow-green. In sufficient amounts the hydrochloride is visible as a greenyellow spot in visible light. Fumes of ammonia reverse this change. This simple test serves to distinguish the alkaloid from a number of naturally occurring fluorescent substances such as methyl anthranilate, anthranilic acid, terthienyl, various coumarins, flavanoids, and alkaloids. il characteristic property of this alkaloid is the marked decrease in fluorescence which occurs in its solution in petroleum ether on the addition of various oxygen-containing solvents such as methanol, ethanol, ’ n-propanol, ether, and acetone. Ethanol is remarkably effective in this respect. However its presence does not significantly alter the ultraviolet absorption spectrum as obtained with a petroleum ether solution of the alkaloid. The possibility that the quenching phenomenon is due to the high dielectric constants of the materials used is unlikely since a concentration of 0.12M of methanol is sufficient to cause a 50% loss of fluorescence in petroleum ether. Workers in Kew Zealand8-12 have isolated and characterized an alkaloid from the aerial portion of Lolium perenne (perennial rye grass) which they have named, perloline. Following this precedent we offer the name “annuloline” for the alkaloid isolated from the roots of the annual rye grass. Perloline and annuloline are clearly different compounds on the basis of their chemical and physical properties, analyses, fluorescent colors, absorption spectra, (7) T. Swain, Biochem. J., 53, 200 (1953). ( 8 ) R. E. R. Grimmet and J. Melville, N e w Zealand J . Sci. Technol., 24B, 149 (1943). (9) R. E. R. Grimmet and D. F. Waters, iVew Zealand J . Sci. Technol., 24B, 151 (1943) (10) I. Ileifer and S. 0. Bathurst, N e w Zealand J . Sci. Technol., 24B, I55 (1943). (11) F. B. Shorland, 9 e w Zealand J . Sci. Technol., 24B, 150 (1943). (12) E.M.Clare and I. M.RIorice, Xeto Zealand J . Sci. Technol., 27B, 36 (1945).
920
NOTES
and distribution. We have been unable to detect any artnuloline in plants of Lolium perenne.
VOL.
23
Anal. Calcd. for C20Hl&oa~C~H3r\;301: C, 55.10; H, 3.89; N, 9.89. Found: C, 55.18; H, 3.86; N, 10.05. ISD. LAFAYETTE,
EXPERIMENTAL
Annzdolarie hydrochloride. Seed of Lolium multi8orum were The Neopentyl and Neophyl Systems in germinated in lots of 350 grams on 47 X 51 cm. sheets of Whatman No. 1 filter paper. Water was supplied daily. Peracid Oxidation of Ketones1 The germination temperature varied between 22-25". After 17 days the aerial portions of the seedlings were discarded JAMES Lv. LI'ILT A N D ALBERT DAXIELZADEH and the roots together with the filter paper in which they were matted were extracted with petroleum ether (b.p. 62"65"). Extraction was performed using 7-10 1. of solvent for Received December SO, S9.V each lot of roots and allowed to proceed for a t least 24 hr. (Kith occasional stirring). The extraction was repeated 2-3 The suggestion of Emmoiis and Lucas2 that the times with fresh solvent. The combined extracts from 2-3 migrating group in the Baeyer-Villiger ketone lots of roots were concentrated to about 200-300 ml. Bfter being stored for several days a t -20" the concentrate was oxidation3 normally is the one most stabilized by filtered through glass wool in order to remove the waxy hyperconjugative electron release appears to be solid which deposited. The concentrate was further reduced substantiated by their work. Thus, in all the alkyl in volume to about 50 ml. and treated with anhydrous HCl whereupon crude annuloline hydrochloride was obtained methyl ketones they reported, the methyl group was as a yellow-gray ppt. The hydrochloride was triturated with never observed to migrate, while the other alkyl about 5 ml. of ice cold absolute ethanol, the mixture centri- group present in each case was capable of such hyfuged a t 2' and the supernatant was discarded. The air perconjugative electron release and gave an dried product was further purified by trituration with 10acetate ester, as indicated below. 25 ml. of HzO and extraction with a sufficient number of 50ml. poitions of petroleum ether to remove the alkaloid mhich a-as now in the form of the free base. The petroleum ether whition \vas concentrated to 50-100 ml. and treated with HC1 gas as before. The hydrochloride was washed with The yields were excellent (mainly 70-9075). We have investigated neopentyl methyl ketone cold absolute ethanol and dried, m.p. 174-177'. 9naZ l 3 Calcd. for C20H20N04Cl: C, 64.30; H, 5.38; N, and neophyl (p-phenylisobutyl) methyl ketone in 3.75; Cl, 9.5. Found: C, 64.89; H, 5.28; i%, 3.57; C1, 8.57, this reaction, following closely the directions of 9.02. Emmons and Lucas. These ketones give neopentyl dnnulolzne. The hydrochloride was converted to the free base by shaking its CHCI, solution with dilute alkali. The and neophyl acetates, as shown belov, with no free base vas obtained as a brown tar on removal of the detectable amounts of methyl or other alkyl esters. CHCl3. Bfter prolonged cooling in an ice bath and stirring, CFGOzOH followed by overnight storage a t room temperature, the CHaCOCH,C(CHa)s CHaCOOCH?C(CH3)3 (2) materiai became transformed into a microcrystalline yellow (C HzCIz) (40%) powder. Recrystallization from hot benzene to which petroleum ether had been added to produce incipient cloudiness yielded fragile, light-yellow narrow rectangles, occurring mainly in clusters; m.p. 105-106O, remelting a t same temperature. The product was neutral and tasteless but CH, CH3 possessing a pleasant spicy resinous odor. The ultraviolet (397) spectrum in cyclohexane showed a maximum a t 354 mp (log * Yield based on consumed starting ketone. E = 4.48)and a minimum a t 285 mp (log E = 3.85). Anal Calcd. for C ~ ~ H ~ O S O ( O CC,H71.20; ~ ) ~ : H , 5.64; X, 4.15: (OCHJ)~,27 6. Found: C, 70.59; H, 5.65; N, 4.16; The yields are poorer, however, thmi those noted in Emmons' work. OCH3, 26.1. The methouyl determination^*^ were made by the titriBecause the neopentyl and neophyl systems have metric method described by Niederl and Xederl 15 Since no hyperconjugative electron release (indeed, these recoverj of OCH, from vanillin standards was 92% of theory, the value shown for annuloline has been adjusted systems often rearrange to the tert-amyl and accordingly. Dioxymethplene groups were shown to be absent bensyldimethylcarbinyl systems in order to achieve by the procedure of Gaebel. '8 such hyperconjugative stabilization), we believe Annuiolzne picrate. A 1yo ethanolic solution of annuloline the present work shows that this hyperconjugative hydrochloride was treated with a slight excess of a 1% ability is not necessary for migration. It seems from ethanolic solution of picric acid. The yellow fibrous crystals all the work reported thus far that any alkyl group n-ere rerrystallized from hot 807, ethanol, m.p. 216-218'. One mole of picric acid per atom of N was found colori- migrates preferentially to methyl. Hyperconjumetrically. gative electron release may, nevertheless, be im~
_
_
(13) Elementarv analyses were performed bv Huffman hlicroanalytical Laboratories, Wheatridge, Colo. (14) The methoxvl and dioxymethylene tests were carried out bv Xlr R. S. Karimoto of this laboratory. (15) J. B. Niederl and V. Niederl, Organic Quantitative Micro-analysis, J. Wiley and Sons, Inc., New York (1938), pp. 187-193. (16) C:. 0. Gaebel, Arch. Pharm., 248, 226 (1910).
(1) Abstracted from the thesis of Albert Danielzadeh to be submitted to the Graduate School of Loyola University for the degree of Master of Science, February 1958. 12) W. D. Emmons and G. B. Lucas, J . rim. Chem. Soe., 77,2287 (1955). 13) A. Baever and V. Villiper. Bcr.. 32. 3625 (1899) and other referenies to be found in the articlh cited in the previous footnote.
